1. Field of the Invention
The invention relates to WLAN (Wireless Local Area Network) communications devices and corresponding methods and in particular to the operation of dual band WLAN communications devices that operate at a frequency in one of two different frequency bands.
2. Description of the Related Art
A wireless local area network is a flexible data communications system implemented as an extension to or as an alternative for, a wired LAN. Using radio frequency or infrared technology, WLAN systems transmit and receive data over the air, minimizing the need for wired connections. Thus, WLAN systems combine data connectivity with user mobility.
Today, most WLAN systems use spread spectrum technology, a wide-band radio frequency technique developed for use in reliable and secure communication systems. The spread spectrum technology is designed to trade-off bandwidth efficiency for reliability, integrity and security. Two types of spread spectrum radio systems are frequently used: frequency hopping and direct sequence systems.
The standard defining and governing wireless local area networks that operate in the 2.4 GHz spectrum, is the IEEE 802.11 standard. To allow higher data rate transmissions, the standard was extended to 802.11b that allows data rates of 5.5 and 11 Mbps in the 2.4 GHz spectrum. Further extensions exist.
Examples of these extensions are the IEEE 802.11a, 802.11b and 802.11g standards. The 802.11a specification applies to wireless ATM (Asynchronous Transfer Mode) systems and is primarily used in access hubs. 802.11a operates at radio frequencies between 5 GHz and 6 GHz. It uses a modulation scheme known as Orthogonal Frequency Division Multiplexing (OFDM) that makes possible data speeds as high as 54 Mbps, but most commonly, communications take place at 6 Mbps, 12 Mbps, or 24 Mbps. The 802.11b standard uses a modulation method known as Complementary Code Keying (CCK) which allows high data rates and is less susceptible to multi-path to propagation interference. The 802.11g standard can use data rates of up to 54 Mbps in the 2.4 GHz frequency band using OFDM. Since both 802.11g and 802.11b operate in the 2.4 GHz frequency band, they are completely inter-operable. The 802.11g standard defines CCK-OFDM as optional transmit mode that combines the access modes of 802.11a and 802.11b, and which can support transmission rates of up to 22 Mbps.
WLAN receivers, transmitters and transceivers, as well as other data communications devices, usually have a system unit that processes radio frequency (RF) signals. This unit is usually called front end.
Basically, a receiver side front end comprises RF filters, intermediate frequency (IF) filters, multiplexers, demodulators, amplifiers and other circuits that could provide such functions as amplification, filtering, conversion and more. Referring to FIG. 1, the front end usually includes an analog front end 100 which is the analog portion of a circuit, which precedes analog-to-digital conversion. Thus, the analog front end 100 performs some analog signal preprocessing in unit 110 and some other functions as described above, and then outputs the analog signal to an analog-to-digital converter 130. The quantized, i.e. digitized, output signal of the analog-to-digital converter 130 is then supplied to a digital signal processor 140.
As can be seen from FIG. 1, the analog front end 100 of conventional data communications receivers may further have a unit 120 for downconverting the received (and preprocessed) analog signal. Conventionally, RF carriers conveying data by way of some modulation technique are downconverted from the high frequency carrier to some other intermediate frequency through a process called mixing. Following the mixing process, the baseband signal is recovered through some type of demodulation scheme.
Receiver architectures exist where unit 120 has zero-IF and/or low-IF topology. This will now be explained in more detail with reference to FIGS. 2 and 3.
FIG. 2 is a simplified diagram illustrating the zero-IF approach for integrated receivers. In the zero-IF approach, the incoming signal, which is at radio frequency, is converted by mixer 200 directly to baseband (BB). Such direct conversion architectures have simplified filter requirements and can be integrated in a standard silicon process, making this design potentially attractive for wireless applications. However, there may be problems with the DC offset, IQ mismatch and with low frequency noise.
FIG. 3 illustrates the low-IF approach. As can be seen, the low-IF architecture operates at an intermediate frequency close to the baseband (like the zero-IF approach) and can therefore be integrated like the zero-IF circuits. However, there is a second downconverter 330 to convert the IF signals to baseband. Low-IF devices can avoid the problems of DC offset, IQ mismatch and low frequency noise but may require additional image rejection. For this reason, an image rejection unit 320 is added in the low-IF topology.
While FIGS. 1 to 3 have been discussed to refer to the receiver side, the transmitter side may be similarly discussed referring to FIGS. 4 to 6. A transmitter side front end comprises RF filters, IF filters, multiplexers, modulators, amplifiers, and other circuits that may provide such functions as amplification, filtering, conversion and more. Referring to FIG. 4, the front end usually includes a digital front end 400 which is the digital portion of a circuit which precedes digital-to-analog conversion. Thus, the digital front end 400 performs some digital signal processing and then outputs the digital signal to a digital-to-analog converter 410. The converted, i.e., analog, output signal of the digital-to-analog converter 410 is then supplied to an analog front end 420.
As can be seen from FIG. 4, the analog front end 420 may have a unit 430 for upconverting the analog signal received from the digital to analog converter 410. Conventionally, baseband carriers conveying data by way of some modulation technique are upconverted from baseband to some other intermediate frequency through a process called mixing. Following the mixing process, the IF signal is further upconverted to an RF frequency in the desired transmission frequency band and is further processed, e.g., filtered or amplified, in unit 440.
FIG. 5 is a simplified diagram illustrating the zero-IF approach for integrated transmitters, and FIG. 6 illustrates the low-IF approach. As can be seen, the low-IF architecture operates at an intermediate frequency close to the baseband (like the zero-IF approach). Further, there are two upconverters 600 and 610 to convert the baseband frequency signals to intermediate frequency and then from intermediate frequency to the transmission RF frequency. Moreover, an LO-feedthrough cancellation unit 620 is added in the low-IF topology. The zero-IF and low-IF approaches shown in FIGS. 5 and 6 have the same or similar characteristics and problems as discussed above with reference to FIGS. 2 and 3.
Another problem with communications devices that operate in a zero-IF or low-IF approach is that the LO signal frequency for up- and downconversion is at the center of the received/transmitted frequency bands. A VCO (Voltage Controlled Oscillator) frequency synthesizer running at this frequency therefore suffers from VCO pulling which significantly degrades the signal quality.
A conventional LO architecture that provides a signal at an output frequency with reduced pulling effect is described in US 2002/0180538 A1. A VCO generates a first signal having a frequency that is a fraction of the output frequency, and a frequency shifter generates a second signal with a frequency substantially equal to the difference between the VCO frequency and the output frequency. Single-sideband mixers are used to produce output signals at the sum of the VCO frequency and the shifted frequency while suppressing an unwanted sideband at the difference of the two frequencies.
While this technique may be suitable for reducing the pulling effect in conventional communications devices, the architecture may have some disadvantages when being applied to dual band WLAN devices. This is in particular because due to the increased number of component parts, the die size and consequently the manufacturing costs are increased. Further, the conventional techniques suffer from power consumption which is sometimes found to be a severe detriment when designing WLAN devices.